University of Twente Student Theses
Effects of Variable Snapshot Frequency on Object Tracking
Huijskes, Hein (2024) Effects of Variable Snapshot Frequency on Object Tracking.
|
PDF
471kB |
| Abstract: | Various areas within artificial intelligence are increasingly adopting environmentally sustainable practices. Part of this movement involves reducing the energy footprint of algorithms, particularly in applications like object tracking. Current practices in evaluating object tracking typically aim for the maximum possible frame rate, which consumes a considerable amount of energy. A lower frame rate that is still effective can reduce environmental costs by lowering energy consumption from object detection algorithms, while still achieving an effective result. This research aims to introduce varied frame rates for camera-feed-based object detection algorithms. State-of-the-art object detection and object tracking algorithms are discussed to find a suitable algorithm for variable frame rate. Together with an in-the-field company a prototype for assessing variable frame rate-based object detection is built and tested using the discussed algorithms. This prototype shows at what lower frame rates object tracking could still be used to adequately detect objects. Different metrics measuring the accuracy of object tracking are compared at different frame rates. This has the potential to minimize hardware and software needs, and electricity consumption. |
| Item Type: | Essay (Bachelor) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 54 computer science |
| Programme: | Computer Science BSc (56964) |
| Link to this item: | https://purl.utwente.nl/essays/101838 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page